We invite you to join us on August 26-27 for a workshop for current and potential users and developers of the Scenic
Scenic probabilistic programming language for world modeling and data generation for AI-based autonomy.
The workshop will be hosted at UC Santa Cruz, and will include food and housing
from dinner on August 25 through
lunch on August 28. The two main days of the workshop (26th and 27th) will provide hands-on tutorials on the
Scenic language, how to interface it to your simulator of choice, and Scenic's use cases for testing, debugging,
synthetic data generation, etc. We will also have breakout groups on topics of special interest such as Scenic's
internals and application domains including autonomous driving and robotics.
Autonomous Vehicles
Robotics
Aviation
Reinforcement Learning
Augmented Reality
Today's autonomous systems rely heavily on the use of machine learning components trained on large amounts of data. Even so, it is expensive to collect relevant data and test these systems in the real world in a manner that captures typical data distributions and also covers edge cases. Therefore, simulators are widely adopted in the robotics and computer vision community to train, test, and debug autonomous and semi-autonomous systems. However, working directly with simulators can be too low-level and problem-specific. To support the design lifecycle of autonomous/semi-autonomous systems, one needs to raise the level of abstraction above individual simulators and provide a formal framework for world modeling. Such a world model can help reason about the safety of a system and facilitate data generation and sim-to-real validation, as well as help to interpret, validate, share, or re-use training and test scenarios across the community.
The objective of this tutorial is to introduce Scenic, an open-source, domain-specific probabilistic programming language for world modeling that addresses the above needs. Scenic is designed to model and generate interactive (or reactive), multi-agent scenarios in a manner portable to any simulator. In Scenic, users can precisely model a stochastic environment in which an autonomous/semi-autonomous system operates, can perform a variety of design and analysis tasks, and can communicate them as interpretable programs. Scenic has a variety of demonstrated use cases, including synthetic data generation, data augmentation, debugging and retraining and redesign of perception components, sim-to-real validation, testing safety of autonomous system both in simulation and in the real world, training reinforcement learning agents in multiplayer settings, and more. To achieve these goals, Scenic has been designed to be (i) intuitive to learn, (ii) probabilistic to capture the uncertainty and stochasticity in the real world, (iii) simulator-agnostic, and (iv) open-source and in the public domain for external members to contribute.
Virtual Attendance:
In-Person Attendance:
8:00 am - 8:00 pm: Check in for parking permit, meal card, dorm room key (West Conference Office)
6:30 pm - 7:15 pm: Dinner in Rachel Carson Dining Hall
8:00 am - 8:00 pm: Check in/out for parking permit, meal card, dorm room key (West Conference Office)
8:15 am: Breakfast in Rachel Carson Dining Hall
9:00 am: Opening, with introductions from participants (Rachel Carson College Room 250)
9:30 am: Introduction to Scenic, tutorial on basic features for static scenarios (Colab Notebook)
10:30 am: Coffee break
11:00 am: Tutorial on Scenic for dynamic scenarios (Colab Notebook)
12:00 pm: Scenic installation help
12:30 pm: Lunch at Rachel Carson Dining Hall
1:45 pm: Hands-on lab session 1: Writing Scenic scenarios
3:15 pm: Coffee break
3:45 pm: Hands-on lab session 2: Scenic API & simulator interfaces (Colab Notebook)
5:15 pm: Free time / chat with the Scenic team
6:30 pm: Dinner at Rachel Carson Dining Hall
7:30 pm: Reception and “Crazy Ideas” Session (Rachel Carson College Room 201)
8:00 am - 8:00 pm: Check out (West Conference Office)
8:15 am: Breakfast in Rachel Carson Dining Hall
9:00 am: Opening, with Scenic case studies (Rachel Carson College Room 250)
9:30 am: Parallel talks/discussions on application domains: Autonomous Driving (in main room); Robotics and Other Domains (in breakout room)
10:30 am: Coffee break
11:00 am: Overview of Scenic's internals
12:30 pm: Lunch at Rachel Carson Dining Hall
1:45 pm: Hands-on session on Scenic development topics
2:30 pm: Parallel hands-on sessions: Synthetic sensor data and VerifAI (main room); Analysis of Scenic programs (in breakout room)
3:15 pm: Coffee break
3:45 pm: Parallel Scenic working group meetings: Autonomous Driving / Intelligent Transportation (main room); Robotics and Other Domains (breakout room); Language evolution and infrastructure (other breakout room).
4:45 pm: Plenary closing session
5:15 pm: Main workshop ends
5:30 pm: Free time / "Scenic" hike around campus
6:30 pm: Dinner at College Nine/Lewis Dining Hall (note different location! hike will end there; shuttle also available)
8:00 am - 8:00 pm: Check out (West Conference Office)
8:15 am: Breakfast in College Nine/Lewis Dining Hall (note different location)
9:30 am: Scenic project next steps discussion Room 250)
10:30 am: Coffee break
11:00 am: Free time / talk with the Scenic team
12:30 am: Lunch at College Nine/Lewis Dining Hall (note different location)
For details, please contact Daniel Fremont.
Website template by Oriane Siméoni from the
Object
localization
for free CVPR '23 tutorial.